On Tue, Nov 1, 2011 at 10:12 AM, Alexander Polakov <polac...@gmail.com> wrote: > * Leroy van Engelen <leroy.vanenge...@gmail.com> [111019 19:07]: >> This was also seen on a macbook by Jan Stary: >> http://marc.info/?l=openbsd-misc&m=131213545109050&w=2 >> >> And on my Samsung N210: >> http://marc.info/?l=openbsd-misc&m=131193104030288&w=2 >> >> I still have this problem, and ran out of options to investigate. The funny >> thing is that, just like the MacBook case above, the high interrupt load >> goes away every other suspend/resume. Do you see this as well? It seems >> like a clue, but I have no idea where to begin investigating, except for the >> ipi code you wrote the diff for. >> > > Hi, > mikeb@ just committed a diff for ppb which solves the problem for me. > Is it the case for you?
Yes, the high interrupt load is gone! However, right after booting the new kernel I saw some weird behaviour. Before rebooting, suddenly my laptop would not resume anymore (can't be related to the fix, but still weird) and I had to use the power switch for a reboot. Then I booted the new kernel, and during boot the system started to fsck the root partition. When the check for the next partition started, it suddonly suspended and when resumed, it would continue for a couple of seconds and resume again. Then, I rebooted the old kernel, let fsck finish without problems and rebooted into the new kernel again. I haven't had the change to test the stability further. If I find out more, I'll post the results. Just out of curiosity, what was the problem and how did you debug it? Bye, -Leroy
